Trends and Advances in Monte Carlo Sampling Algorithms

Application Deadline for this workshop was October 20, 2017

Location

This workshop will be held at the Penn Pavilion (Level 2), Duke University, Durham, NC.

Description

This second workshop in the SAMSI Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applied Mathematics focusses on Monte Carlo sampling methods –an important class of computational algorithms for estimating high dimensional distributions. Monte Carlo sampling is widely used in physics, chemistry, mathematics and statistics, and is most useful when other methods fail due to the high dimensionality of the problem. Due to the extensive application of Monte Carlo sampling across disciplines, breakthroughs in one discipline can lead to advances in others.

This SAMSI workshop brings together experts from applied mathematics, statistics and machine learning for the purpose of exchanging ideas and advancing the broad area of sampling algorithms.


Schedule and Supporting Media

Printable Schedule
Printable Map
Participant List

Lecturers

Confirmed Speakers currently include:

Monday, December 11, 2017
Penn Pavilion, West Campus, Duke University

Time Description Speaker Slides Videos
8:30-9:00am Registration
9:00-9:15am Introductions and Welcome Ilse Ipsen, SAMSI Associate Director
9:15-9:55am Component-wise Markov Chain Monte Carlo Galin Jones, University of Minnesota
9:55-10:15am Discussion
10:15-10:55am Approximate MCMC in Theory and Practice James Johndrow, Stanford University
10:55-11:15am Discussion
11:15-11:45am BREAK
11:45am-12:25pm Sparse Polynomial Approximation via Compressed Sensing of High Dimension Functions Hoang Tran, Oak Ridge National Laboratory
12:25-12:45pm Discussion
12:45-2:00pm LUNCH
2:00-2:40pm Principled Variational Learning and Inference for Deep Generative Neural Networks Larry Carin, Duke University
2:40-3:00pm Discussion
3:00-3:40pm Stochastic Gradient MCMC for Independent and Correlated Data Yian Ma, University of Washington
3:40-4:00pm Discussion
4:00-4:30pm BREAK
4:30-5:10pm Parallel Markov Chain Monte Carlo Scott Schmidler, Duke University
5:10-5:30pm Discussion
5:30 Shuttle to Hotel

Tuesday, December 12, 2017
Penn Pavilion, West Campus, Duke University

Time Description Speaker Slides Videos
8:45-9:00am Registration
9:00-9:40am Multiscale Implementation of Infinite-Swap Replica Exchange Molecular Dynamics Eric Vanden-Eijnden, New York University
9:40-10:00am Discussion
10:00-10:40am Lecture to be Announced Jonathan Mattingly, Duke University
10:40-11:00am Discussion
11:00-11:30am BREAK
11:30am-12:10pm Discontinuous Hamiltonian Monte Carlo for Sampling Discrete Parameters Akihiko Nishimura, University of California, Los Angeles
12:10-12:30pm Discussion
12:30-2:00pm LUNCH
2:00-2:40pm Measuring Sample Quality with Kernels Lester Mackey, Microsoft Research
2:40-3:00pm Discussion
3:00-3:40pm A Stein Variational Approach for Deep Probabilistic Modeling Qiang Liu, Dartmouth College
3:40-4:00pm Discussion
4:00-4:30pm BREAK
4:30-5:10pm When to Stop Sampling: Answers and Further Questions Fred Hickernell, Illinois Institute of Technology
5:10-5:30pm Discussion
5:30-7:00pm Poster Session and Reception
7:00pm Shuttle to Hotel

Wednesday, December 13, 2017
Penn Pavilion, West Campus, Duke University

Time Description Speaker Slides Videos
8:45-9:00am Registration
9:00-9:40am Variance Reduction via Taylor Approximation of High-dimensional Parameter-to-output
Maps Governed by Expensive-to-solve PDEs: Applications to optimal control under uncertainty
Omar Ghattas, University of Texas
9:40-10:00am Discussion
10:00-10:40am Approximate Bayesian Computation for Mechanistic Network Models Antonietta Mira, Università della Svizzera Italiana and Università dell’Insubria
10:40-11:00am Discussion
11:00-11:30am BREAK
11:30am-12:10pm Self-adjusted Mixture Sampling and Locally Weighted Histogram Analysis Methods Zhiqiang Tan, Rutgers University
12:10-12:30pm Discussion
12:30-2:00pm LUNCH
2:00-4:00pm Open Discussion
4:00-4:30pm BREAK
4:30-5:10pm Discussions/Collaborations
5:10pm Shuttle to Hotel

Thursday, December 14, 2017
Penn Pavilion, West Campus, Duke University

Time Description Speaker Slides Videos
8:45-9:00am Registration
9:00-9:40am Sequential Inference via Low-dimensional Couplings Youssef Marzouk, MIT
9:40-10:00am Discussion
10:00-10:40am About Infinite-Dimensional Geometric MCMC Shiwei Lan, Caltech
10:40-11:00am Discussion
11:00-11:30am BREAK
11:30am-12:10pm Automated Scalable Bayesian Inference via Data Summarization Tamara Broderick, MIT
12:10-12:30pm Discussion
12:30-2:00pm LUNCH
2:00-2:40pm Expediting Monte Carlo Sampling via Multi-fidelity Information Fusion Paris Perdikaris, MIT
2:40-3:00pm Discussion
3:00-3:40pm Regularization and Computation with High-dimensional Spike-and-slab Posterior Distributions Yves Atchade, University of Michigan
3:40-4:00pm Discussion
4:00-4:30pm BREAK
4:30-5:10pm Jittered Sampling: Bounds and Problems Stefan Steinerberger, Yale University
5:10-5:30pm Discussion
5:30pm Shuttle to Hotel

Friday, December 15, 2017
Penn Pavilion, West Campus, Duke University

Time Description Speaker Slides Videos
8:45-9:00am Registration
9:00-9:40am Importance Sampling the Union of Rare Events with an Application to Power Systems Analysis Art Owen, Stanford University
9:40-10:00am Discussion
10:00-10:40am Stratification for Markov Chain Monte Carlo Simulation Jonathan Weare, University of Chicago
10:40-11:00am Discussion
11:00-11:30am BREAK
11:30am-12:10pm Robust MCMC Sampling with Non-Gaussian and Hierarchical Priors in High Dimensions Matthew Dunlop, Caltech
12:10-12:45pm Discussion and Wrap-Up
12:45pm Shuttle to RDU Airport

Questions: email qmc@samsi.info